Oral Maxillofac Surg (2012) 16:1927 DOI 10.

1007/s10006-011-0287-4

REVIEW ARTICLE

Risk factors and prevention of bad splits during sagittal split osteotomy
Bruno Ramos Chrcanovic & Belini Freire-Maia

Received: 15 January 2011 / Accepted: 3 August 2011 / Published online: 12 August 2011 # Springer-Verlag 2011

Abstract Purpose One of the operative complications of the sagittal split osteotomy of the mandible is a bad split, which describes an unfavorable or irregular fracture of the mandible in the course of the osteotomy. The purpose of this study is to identify previous studies which reported incidences of bad split occurrence during sagittal split osteotomy and to discuss its mechanisms and risk factors, based on a literature review, in order to minimize their occurrence. A few illustrative cases are also presented. Methods An electronic search was undertaken in January 2011. The titles and abstracts from these results (n =363) were read for identifying studies which reported incidences of bad split occurrence during sagittal split osteotomy procedures. Results Twenty-one studies were identified and assessed. The incidence of bad splits from these studies varied between 0.21% and 22.72%. The buccal plate of the proximal segment and the posterior aspect of the distal segment were the most affected areas. Discussion The surgical patient should be evaluated according to age and the presence of unerupted/impacted

third molars. Prevention is focused on adequate osteotomy design, eliminating sharp angle where abnormal stress occurs on bony segments, completion of adequate cuts into the retrolingular depression and through the inferior border, and careful separation of the segments. The SSO is an extremely technical and sensitive procedure, and careful attention will probably prevent most unfavorable splits. If a fracture occurs, the fractured segments should be incorporated into the fixation scheme if possible. The occurrence of bad splits cannot always be avoided. When adequately treated the chances of functional success are good. Keywords Orthognathic surgery . Sagittal split osteotomy . Bad split . Complications . Third molar

Introduction Sagittal split osteotomy (SSO) of the mandible is one of the most common operative techniques used in orthognathic surgery. It is a very versatile surgical procedure to advance or set back the mandible. Since its initial description by Schuchart [1], various modifications have been advocated by Trauner and Obwegeser [2], Mathis [3], Dal Pont [4], Hunsuck [5], Epker [6], Wolford et al. [7], Loh [8], and Marquez and Stella [9] to decrease the incidence of its complications. Despite all improvements and being commonly used, it requires a precise surgical technique and might not be free of complications [4, 5, 10], the most common being unwanted fracture of either the distal or proximal segment, damage to the inferior alveolar nerve, bleeding, inadequate fixation, and relapse [1013]. One of the operative complications during SSO is a bad split, which describes an unfavorable or irregular fracture of the mandible in the course of the SSO [11, 14, 15]. Bad

B. R. Chrcanovic (*) Av. Raja Gabaglia 1000/1209, Gutierrez, Belo Horizonte( MG CEP 30441-070, Brazil e-mail: brunochrcanovic@hotmail.com B. Freire-Maia Av. do Contorno 4747/1010, Serra, Belo Horizonte( MG-CEP 30110-921, Brazil e-mail: belinimaia@lifecenter.com.br B. R. Chrcanovic : B. Freire-Maia Department of Oral and Maxillofacial Surgery, School of Dentistry, Pontifcia Universidade Catlica De Minas Gerais, Belo Horizonte, Brazil

20

Oral Maxillofac Surg (2012) 16:1927

splits during SSO can assume several forms and occur in different places, such as: on the superior corner of the proximal segment, inferior border left intact on the distal or proximal segment, nerve trapped in the proximal segment, inadvertent cut through the ramus, distal segment vertical fracture, and buccal plate or proximal segment fracture [10, 16]. The bad split involving the condylar process is the most difficult to treat, especially when it occurs posterior to the proximal segment and the condylar process is left attached to the distal fragment [13]. Bad splits can be provoked by an anatomically thin mandibular ramus, a high mandibular lingula, the presence of third molars (3M), incorrect inclination of the osteotome, or even by the inexperience or lack of attention of the surgeon [13, 17, 18]. Some authors suggest that the use of heavy osteotomies and twisting techniques could be the main cause of bad splits [19], or that the method and instruments used for initiating and completing the split are importantchisel vs. spreader, fine chisels vs. heavy chisels, one hit vs. more than one hit, and position and sequence of the osteotomes used [18]. Others think that the incomplete split of the inferior border of the mandible could cause this complication [20]. There is no consensus in the literature as to what combination of factors predisposes to a bad split [13]. A bad split can lead to infection, bony fragment sequestration, delayed bone healing, and pseudoarthrosis [11, 15, 16]. Postoperative instability, relapse, or dysfunction of the mandible with consecutive impairment of the temporomandibular joint may develop [21]. Postoperative instability or dysfunction of the mandible with consecutive TMJ dysfunction may occur [15]. Opinions differ about the removal of the 3M before or during the SSO, because the presence of an unerupted 3M tooth during surgery has been proposed as a factor that increases the risk of unfavorable split [14, 15, 22, 23]. Older patients also seem to have a greater tendency to suffer a bad split [21]. As these operations are usually elective procedures and in some cases only for esthetic purposes, knowledge of the potential risks is essential for the surgeon, orthodontist, and patient. Furthermore, it is crucial to understand the mechanism of complications to minimize potential risks [10]. The purpose of this study is to identify previous studies which reported incidences of bad split occurrence during SSO and to discuss its mechanisms and risk factors, based on a literature review, in order to minimize their occurrence. A few illustrative cases are also presented.

National Library of Medicine, National Institutes of Health). The search terms (a) sagittal split osteotomy AND complication were used in the first search performed, and then (b) sagittal split osteotomy AND complications, (c) sagittal split osteotomy AND bad split, (d) sagittal split osteotomy AND unfavorable split, (e) orthognathic AND complication, and (f) facial orthopedic surgery AND complication. Reference lists of the included studies were also screened for potentially relevant research. For each of the identified articles included in this study, the following data were then extracted on a standard form: year of publication, number of patients who underwent SSO, number of SSO performed, occurrence of bad splits, incidence of bad splits per patient and per SSO (in percentage of the entire sample), and reported incidences of proximal and distal segment fractures. Search strategy development, study selection, and data collection were done by one author (B.R.C.). Based on a review of these articles, the following focused questions were raised and will be discussed: What is the incidence of bad splits? Which segments of bone are the most affected? Does the presence of 3M or the age of the patient influence the incidence of bad splits? 3Ms should be removed some months before SSO or not? What to do when a bad split happens?

Results The searches resulted in 29 citations from (a), 211 from (b), eight from (c), 12 from (d), 68 from (e), and 35 from (f). The titles and abstracts (when available) from these results (n = 363) were read for identifying studies which reported incidences of bad split occurrence during SSO procedures. Twenty-two citations were identified with this criterion (a=4, b=8, c=2, d=3, e=4, f=1); seven were cited in more than one research of terms (one study was cited in three research of terms). Thus, 14 studies were identified without repetition. Seven more studies were found from citations of these first 14 articles, giving a total of 21 studies (Table 1). The incidence of bad splits from 21 studies varied between 0.21% and 22.72% (Table 1) [1014, 1618, 21, 22, 2434]. It can be observed that because most SSOs are done bilaterally and a bad split only rarely occurs as a bilateral complication, the reported incidence nearly doubles when patients, not sites, are used as a counting base [34]. The buccal plate of the proximal segment and the posterior aspect of the distal segment were the most affected areas. Some authors have reported incidences of proximal segment fracture as follows: Guernsey and DeChamplain [11] 4.5%, Jnsson et al. [24] 8.9%, MacIntosh [25] 3.3%, Martis [17] 0.78%, Turvey [14]

Materials and methods An electronic search without date or language restrictions was undertaken in January 2011 in PubMed website (US

Table 1 Reported incidences of bad split occurrence during SSO procedures

Occurrence Incidence per Incidence per Patients age range (years) Patients mean age patient (%) SSO (%) (years) Surgical technique 5 10 5 16 5 9 5 21 97 24 24.4 NM 23g NM 17.7 (G1) 36.6 (G2) 30.3 18.5 (G1) 27.3 (G2) 23.1 NM 24 6 8 34 11 12 4 12 9 12 12 14 1.39 0.70 10.91 5.45 10.91 5.45 3.85 1.92 f 0.95 1453 5.71 2.88 1349 2.33 1.17 1560 4.20 2.20 1344 (G1) 1756 (G2) 10.12 5.06 NM 1.00 0.50 1373.9g 0.21 NM 3.79 1.91 12.157.9 7.87 3.93 f 10.14 5.25 NM NM 4.03 25.1 1559 7.03 3.56 656 (G1) 1246 (G2) 25.2 (G1) 26.2 (G2) NM Obwegeser-Dal Pont [4]+Hunsuck [5] Duguet et al. [46] Hunsuck [5] Own technique and Dal PontHunsuck-Simpson-Epker [4, 5, 47] Obwegeser-Dal Pont [4] Wolford et al. [7] NM NM Obwegeser [2] NM 26 (G1) 22 (G2) 26 (G3) Obwegeser-Dal Pont [4] 32.7 (G1) 24.4 (G2) 1561 26.0 Obwegeser-Dal Pont [4] Epker [6] 1.94 0.97 1440 20 6.78 25.0 1353 17.86 8.93 NM NM 1.67 NM NM Obwegeser [2] Obwegeser-Dal Pont [4] Obwegeser-Dal Pont [4] Obwegeser-Dal Pont [4] Obwegeser-Dal Pont [4] Obwegeser-Dal Pont [4] 22.72 11.36 15.532 21.9 Obwegeser [2]

Authors

Published Patients SSO

Guernsey and DeChamplain [11] 1971 600 28 236 258 128 124 400 2,466 1,256 2,820 1,584 672 500 1,030 139 468 220 220 2,005 207 1,233 633 802 336 262 515 70 1,264 234 110 110 1,008 256 516 56

22

44

Behrman [12]

1972

Jnsson et al. [24]

1979

MacIntosh [25]

1981

Martis [17]

1984

Oral Maxillofac Surg (2012) 16:1927

Turvey [14]a

1985

Van Merkesteyn et al. [22]

1987

Tucker [26]

1995

Van de Perre et al. [27]

1996

Precious et al. [28]

1998

Akhtar and Tuinzing [18]

1999

Acebal-Bianco et al. [29]

2000

Maurer et al. [30]

2001

Mehra et al. [21]b

2001

Panula et al. [16]

2001

Reyneke et al. [31]c

2002

Teltzrow et al. [10]

2005

Kim and Park [32]

2007

Kriwalsky et al. [33]

2008

Veras et al. [13]e

2008

1760 (G1) 1739 (G2) 1760 (G3) 2543 (G1) 2238 (G2)

Falter et al. [34]

2010

NM not mentioned

Turvey [14] divided their patients into two groups: group 1 consisted of 64 patients operated by the author and three senior residents. These patients underwent sagittal osteotomies that extended to the second or first molar regions of the body of the mandible. Group II consisted of 64 patients operated by the author and a different group of three senior residents. In group II patients, the osteotomy was confined to the retromolar region of the mandible except in three patients who underwent advancement greater than 10 mm

Mehra et al. [21] divided their patients into two groups: group 1 included patients who underwent concomitant removal of third molars at the time of surgery and group 2 included patients with either congenital absence of third molars or who had the third molars removed at least 1 year before the SSO was performed

G1age of the group with unerupted third molar teeth present; G2with third molar teeth absent

Kriwalsky et al. [33] divided their patients into three groups: group 1sagittal split where the third molar was absent, or at least 6 months after removal of the tooth; group 2sagittal split with a retained or impacted third molar, which was removed simultaneously; and group 3sagittal split with a third molar present, which was left in place

Veras et al. [13] divided their patients into two groups: group 1 included patients with bad splits, and group 2 without bad splits (normal split cases)

Van de Perre et al. [27] and Kim and Park [32] reported the age distribution of the patients and the information of patients age range was not available

Data concerning all the patients, also involving maxillary and chin surgery 21

22

Oral Maxillofac Surg (2012) 16:1927

3.1%, Van Merkesteyn et al. [22] 4%, Precious et al. [28] 1.2%, Teltzrow et al. [10] 0.47%, and Kriwalsky et al. [33] 5.45%. The incidence of distal segment fracture has been reported by Guernsey and DeChamplain [11] as 6.8%, Martis [17] 0.19%, Turvey [14] 0.4%, Precious et al. [28] 0.72%, and Teltzrow et al. [10] 0.08%. No distal fracture occurred in studies conducted by Jnsson et al. [24], MacIntosh [25], Van Merkesteyn et al. [22], or Kriwalsky et al. [33].

Discussion As noted in this review of the literature, the incidence of bad splits varied greatly among the 21 studies (between 0.21% and 22.72% Table 1). Some results should be better analyzed in order to understand these large differences in the statistical results. For example, the higher incidence of bad splits in younger patients may be due to larger number in that age group in that study. In the study of Van de Perre et al. [27], 76.5% of all patients were in the 1529 age range. And in the study of Kim and Park [32] 33.2% of the patients are between 15 and 19 years of age, and 62.8% between 20 and 29 years. In addition, there are several SSO modifications, and the same technique was not used in all studies, and the expertise of the surgeons may vary from study to study. However, Falter et al. [34], who examined the clinical records and radiographs of 1,008 consecutive patients who underwent at least one SSO, were unable to demonstrate a significant decrease in incidence of bad splits during the 20-year period of their study; the influence of the growing expertise of the same qualified surgeons could not be established as an influencing factor. They stated that aside from operator experience and skill, complications appear to be related to the osteotomy design and attention to detail during the operation. As the incidence of bad splits during SSO can be considerable, as demonstrated in some of these previous studies, it is important to understand the factors complicating this procedure to minimize this incidence. One of the most reported factors in the literature is the presence of an impacted 3M. There is still controversy about whether the presence of a retained or impacted 3M in the lower jaw increases the risk of a bad split or not [33]. Still it cannot be said for sure whether it is better if it should be removed before orthognathic surgery or during SSO [28]. The decision depends on the surgeons experience, the site, angulation, relative height, and root form of the 3M, and its morphological relation to the neurovascular bundle. When the traditional Obwegeser technique is used, the removal of the 3M may not be necessary, as it is not touched by the osteotomy line and remains uncovered [2]. If it is removed before orthognathic surgery, complete bone

healing and mineralization should be allowed [23]. However, according to Kriwalsky et al. [33] the 3M has no influence on the incidence of atypical sagittal splits. Some authors favor the removal of 3Ms at least 6 months before SSO [15, 18, 23, 34] and others do not [20, 21, 26, 28, 33]. Authors who advocate the removal of impacted 3Ms during the SSO surgery state that when there is aggressive removal of the bone in an impacted 3M surgery, particularly lingually, it can predispose the mandible to an unfavorable fracture during a later SSO surgery, especially in cases where the mandible is already thin or where anatomic irregularities exist [20, 24, 28]. The buccal cortex of the mandible normally is thin posterior to the second molar, weakening this region, and this can contribute to fracture of the proximal segment [35]. In very thin bony structures of the mandibular angle, the complete cutting of the cortical structures is necessary, until bleeding spongy bone in the osteotomy groove can be seen. The procedure requires a careful and slow distraction of the osteotomy gap to prevent sudden wrong fractures in the deperiosted buccal or lingual segment. The presence of 3Ms can further thin and weaken this region [36]. In the presence of 3Ms, the lingual cortex of the distal segment is also usually very thin, and this increases the risk of a vertical fracture through the tooth socket. Removal during SSO allows a better operative view, which facilitates the removal and reduces the risk of injury to the inferior alveolar nerve. A second operation for the patient and the additional loss of bone from the osteotomy are avoided [20]. In well-defined cases the SSO is recommended, even including removal of the impacted 3M to reduce nerve injury [37]. The roots of the wisdom tooth can serve as a good guide for the chisel during osteotomy lateral to the nerve. El Deeb et al. [23] also recommended that 3Ms be removed at the time of the SSO but, if the surgeon prefers to remove them before the SSO, it should be done at least 9 to 12 months before surgery to allow for complete socket bone fill and maturation. According to Reyneke et al. [31], unerupted 3M tooth removal 6 to 9 months before orthognathic surgery is recommended in patients younger than 20 years to reduce the risk of complications, but for those older than 20 years, prior removal may not be necessary. Authors who advocate the removal of 3Ms at least 6 months before SSO claim that a vertical fracture through the 3M socket may occur after uneventful splitting of the mandible, during removal of the impacted 3M (surgically sectioning the impacted tooth and removing it in segments can help prevent this from occurring) [21]. SSO and simultaneous removal of 3M teeth increases operating time and technical difficulty of the osteotomy and fixation

Oral Maxillofac Surg (2012) 16:1927

23

procedures [26]. The presence of an unerupted 3M or an empty tooth socket after removal of the tooth while performing the SSO also influences the positioning of bicortical screw or plate fixation [31]. This can limit the areas for placement of screws or plates (Fig. 1). Unfortunately, the configuration of the screw placement or screws placed in areas of thin bone can lead to poor fixation [31]. The standard removal of unerupted 3Ms in all patients 6 months before SSO might also be an element explaining the low incidence of bad splits in the series of a recent study [34]. This also reveals that after removal of unerupted 3Ms at least 6 months before SSO, no bad splits were found in the group younger than 20 years. These results tend to confirm the statement of Mehra et al. [21] that unfavorable fractures usually occur in young, teenage patients when 3Ms are present during the SSO, and with the results of the study of Reyneke et al. [31], in which all fractures occurred in patients younger than 20 years who had 3M teeth present and removed at the time of surgery. In a retrospective study involving 1,256 SSOs, Precious et al. [28] found that fracture of the split segments occurred with a greater frequency when the impacted 3M teeth were removed at least 6 months before surgery (2.62%) than when unerupted 3M teeth were removed concomitant with the SSO (0.94%). Based on their findings, they suggested that removal of 3Ms at least 6 months before SSO does not reduce the incidence of unfavorable fractures and, in most cases, the 3M removal and SSO can be safely carried out in one operation. However, more precise information is needed and the use of register-based investigations together with well-designed prospective follow-up studies appears to be crucial. One of the major difficulties in evaluating the literature on incidence of bad splits is, indeed, the lack of comparable prospective studies. Almost all studies quoted in Table 1 are retrospective studies. Tucker [26] and

Fig. 1 The blue areas indicate where rigid fixation can be placed, when there is the presence of an unerupted third molar (red area) or empty tooth socket, which limits the areas for placement of screws or plates (after Reyneke et al. [31])

Reyneke et al. [31] were the only researches who conducted a prospective and nonrandomized study about this issue. In the first study [26], involving 400 SSOs, unfavorable fractures occurred in 4% of SSOs with 3Ms present and in 3% with 3Ms absent. It was concluded that the incidence of unfavorable fracture was not significantly different between the two groups. However, in the second study with 139 SSOs [31], all fractures occurred in patients who had 3M teeth present and removed at the time of surgery. Thus, the results of the only two prospective studies seem to not totally agree with this issue. Therefore, more welldesigned prospective studies are necessary. Some authors have provided suggestions for those who even in the presence of 3M decide to do the SSO. The modification of the SSO procedure is accomplished using a spreading versus malleting and chiseling technique [9]. The cut of mandibular basis should be extended properly to prevent unfavorable fracture at the proximal segment and any bone removal overlying the impacted molar must be avoided until the sagittal splitting procedure has been completed [38]. The spreading should be done with spreaders in the superoinferior direction instead of anteroposterior direction because of the resistance of cortical bone under the inferior alveolar channel [9, 38]. Moreover, a modified osteotome that was designed specifically for inferior border osteotomy of the mandible [39] to properly separate the mandible and prevent the bad split of the segments during the SSO. The patients age is also identified in literature as an important influencing factor for the occurrence of bad splits. Advancing age can increase the risk of a bad split and should be considered a complicating factor. In the study of Kriwalsky et al. [33], with an evaluation of a total of 220 SSO of the mandible, the mean age of all patients who had bad splits was 35 (range 2160)years compared with 25 (1745)years in patients who did not. In the study of Veras et al. [13], the mean age of the patient group with bad splits was 32.7 years (2543 years) and 24.4 years in the patient group without bad splits (2238 years). In the study of Falter et al. [34] with 2005 SSO, the average age of the patient group with bad splits was 33.1 years, whereas the mean age of the patient group without bad splits was 25.9 years. However, there is an important factor that contrasts with the data from these studies. Mehra et al. [21] divided their patients into two groups: group 1 included patients who underwent concomitant removal of 3Ms at the time of surgery and group 2 included patients with either congenital absence of 3Ms or who had the 3Ms removed at least 1 year before the SSO was performed. They found that when 3Ms are present during the SSO, unfavorable fractures usually occur in young, teenage patients (group 1average age 15.9 years). In the group 2, two patients with fractures were aged 42 and 44 years.

24

Oral Maxillofac Surg (2012) 16:1927

Reyneke et al. [31] also found that all cases of bad splits occurred in patients with retained or impacted 3Ms, and all patients were under 20 years old. Moreover, this study [31] also showed that the SSO procedure in the group with unerupted 3M teeth was significantly easier in the older age group (mean 22.3 9.6 years) and significantly more difficult in the younger age group (mean, 15.91.6 years). The fact that unerupted 3M teeth in younger patients are positioned lower in the mandible than in older patients may influence the strength of the segment [31]. The softer cortex in younger patients may also make greenstick fracture easier than in older patients [21]. This anatomic fact is most probably the reason for the increased difficulty of the surgical procedure, as well as increased complication rate, in the younger patient [31]. As the young patient contains more spongy bone because of the recently terminated bone growth, the bone may be weaker and softer. Thus, there may be no adequate structure support for the chisel for the separation of the cortical layer. Bone in patients of more than 50 years is no longer dynamic, but more brittle and fragile than in young patients. Thus, the splitting procedure needs more time and concentration, totally complete cortical grooves, and the spreading has to be done as slowly and carefully as possible. Some works have identified more variables. The results of one study [21] showed that the degree of 3M impaction and development did affect the incidence of unfavorable splits. Complete tooth impaction was seen in six of the eight fractures (75%), and five of the eight fractures (62.5%) were associated with two thirds crown formation. Another study [13] observed that patients with bad splits presented a statistically significant smaller mouth opening compared with patients without bad splits (P < 0.05). However, this limitation in mouth opening was suggested most likely to happen in a kind of bad split. It has been reported that excessive manipulation of the proximal segment could cause intra-articular hematoma, leading to pain or limited mouth opening [30]. This was observed by Veras et al. [13]. They observed that most bad splits occurred at the proximal segment and therefore required

Fig. 3 3D-CT, postero-anterior view of the mandible. Bad split of the distal segment (red arrow) on the left mandibular side

extra stabilization and consequently more manipulation, which could have explained the smaller mouth opening in the bad split group. They also observed that no tendency to develop significant temporomandibular disorder was observed in patients with bad split. Moreover, when a bad split happens and more surgery and stripping of periosteum are needed, there might be a higher incidence of swelling, inferior alveolar nerve damage and worse long-term stability. One study showed that post-surgical neurosensory impairments can be indirectly caused by edema due to the compression of the nerve bundle within the mandibular canal [40]. In addition, as a consequence of intraoperative swelling and inflammation within the joint because of too much manipulation of the proximal segment, an increase in vertical joint space is a common finding [41]. Greater manipulation of the bone segments is usually necessary when trying to reposition the fractured segment in the scheme of rigid fixation. The importance of correct positioning of the condyles before fixation is well-known [14]. Improper positioning of the condyle in the glenoid fossa at time of surgery can affect the long-term stability, and consequently cause relapse [41]. The main concern when a bad split happens should be the disturbance in the bony union, which could cause sequestration of the fractured segments with an increase in infection rate. For this reason, some authors recommend intermaxillary fixation in case of a bad split to ensure

Fig. 2 Panoramic radiography. Unfavorable fracture of the proximal segment on the left side (L) and favorable split on the right side (R)

Fig. 4 Same case presented in Fig. 3 in axial CT

Oral Maxillofac Surg (2012) 16:1927 Fig. 5 ac 3D-CT. Fracture of the posterior border of the mandibular ramus

25

correct bone healing [20]. However, in a study where the functional and radiographic long-term results after bad split in orthognathic surgery were evaluated [13], small buccal plate bad splits do not usually undergo any change on the post-surgical treatment plan and some authors emphasize that this sort of split does not lead to functional interference. This fact was confirmed by a recent study. According with the results of this study [34], even if a bad split occurs, this has no influence on the postoperative course or the end result. All bad splits could be easily repaired by additional osteosynthesis measures resulting in sufficient rigid skeletal fixation, not necessitating postoperative intermaxillary fixation [34]. If a fracture occurs, fractured segments should be reconsolidated if possible. Fractured segments should be incorporated into the fixation scheme to avoid unfavorable post-surgical positional changes and provide stable continuity between the most proximal portion of the mandible and the distal segment [13, 42]. Resection of the coronoid process in order to use it as a free cortical graft was also suggested [43]. Patterson and Bagby [44] discussed fracture of the proximal segment and noted that completion of the split

can be extremely difficult. The authors recommended lateral retraction of the remaining proximal segment and vertical osteotomy of the distal segment to complete the split [44]. However, according to ORyan and Poor [45], this procedure may limit the amount of advancement and may place the lingual nerve at risk from stretching or retraction of the medial tissues. These latter authors recommended that if the buccal plate has been fractured, it should be plated immediately using a four-hole plate with 5 mm unicortical screws. And then use a spreader with moderate force to complete the split as though the buccal plate were intact. According to them [45], this method is best applied when the buccal plate fracture is anterior to the mandibular angle and should not be used when the fracture is near the condyle. They also reported that this technique was used in five patients, all of whom healed uneventfully. Some illustrative cases are shown with the following description. Figure 2 shows a postoperative panoramic radiography after bilateral SSO. Note the unfavorable fracture of the proximal segment on the left side (L), and compare the favorable split on the right side (R). Rigid fixation with an additional miniplate was required for stability of the segments.

Fig. 6 Same case presented in Fig. 5 in coronal CT

Fig. 7 Same case presented in Fig. 5 in axial CT

26

Oral Maxillofac Surg (2012) 16:1927

Figure 3 shows a three-dimensional computed tomography (3D-CT), postero-anterior view of the mandible, with bad split of the distal segment on the left side of the mandible (red arrow). Figure 4 demonstrates the same case presented in Fig. 3, but in axial TC. Figure 5 shows fracture of the posterior border of the mandibular ramus in three different view angles in 3D-CT. The same case presented in Fig. 5 is showed in coronal CT (Fig. 6) and in axial CT (Fig. 7).

the retrolingular depression and through the inferior border, and careful separation of the segments. The SSO is an extremely technical and sensitive procedure, and careful attention will probably prevent most unfavorable splits. If a fracture occurs, fractured segments should be reconsolidated if possible. As a final remark, the occurrence of bad splits cannot always be avoided. When adequately treated the chances of functional success are good.

Conclusions Three main conclusions can be drawn from this review. They are also suggestions/measures to try to prevent the occurrence of bad splits. First, the surgical patient should be evaluated according to age and the presence of unerupted/impacted 3Ms. Young teenagers have much thinner, immature cortical bone than adults and, with additional thinning caused by development of the coronal portion of the 3M. This patient population may have a greater predisposition to unfavorable fractures. Thus, greater care must be taken during surgery in the young teenage patient with 3Ms to avoid unfavorable splits. In very thin bony structures of the mandibular angle, the complete cutting of the cortical structures is necessary, until bleeding spongy bone in the osteotomy groove can be seen. The procedure requires a careful and slow distraction of the osteotomy gap to prevent sudden wrong fractures in the deperiosted buccal or lingual segment. Second, unerupted 3M tooth removal 6 to 9 months before orthognathic surgery is recommended in patients younger than 20 years to reduce the risk of complications, but for those older than 20 years, prior removal may not be necessary. However, some authors favor the removal of 3Ms at least 6 months before SSO and others do not, making this subject still a controversial issue. The feasibility of removing unerupted 3Ms during the SSO should be evaluated in a case by case basis. Advantages such as elimination of a second surgical procedure and removal of a possible intraosseous interference during approximation of the distal and proximal segments should be considered. However, the risk of distal segment fracture and/or inferior alveolar nerve damage during removal should also be considered. Third, risk factors should be identified and reduced so far as possible, particularly because it is an elective operation. All the modifications to the technique were developed to facilitate the procedure and assure a safe split and they have their own, sometimes very specific, indications. The surgeon must decide on the best technique for each case. Prevention is focused on adequate osteotomy design, eliminating sharp angle where abnormal stress occurs on bony segments, completion of adequate cuts into
Conflict of interest The authors declare that they have no conflict of interest.

References
1. Schuchart K (1942) Ein Beitrag zur chirurgischen Kieferorthopadie unter Berucksichtigung ihrer Bedeutung fur die Behandlung angeborener under worbener Kieferdeformitaten die Soldaten. Dtsch Zahn- Mund-Kieferhk 9:7389 2. Trauner R, Obwegeser H (1955) Zur Operationstechnik bei der Progenie und anderen Unterkieferanomalien. Dtsch Zahn Mund Kieferheilk 23:1126 3. Mathis H (1956) ber die Mglichkeit der rein enoralen Durchfhrung der beiderseitigen Osteotomie zur Behandlung der Progenie. sterr Z Stomat 53:362 4. Dal Pont G (1961) Retromolar osteotomy for the correction of prognathism. J Oral Surg Anesth Hosp Dent Serv 19:4247 5. Hunsuck EE (1968) A modified intraoral sagittal splitting technique for correction of mandibular prognathism. J Oral Maxillofac Surg 26:250253 6. Epker BN (1977) Modifications in the sagittal osteotomy of the mandible. J Oral Surg 35:157159 7. Wolford LM, Bennett MA, Rafferty CG (1987) Modification of the mandibular ramus sagittal split osteotomy. Oral Surg Oral Med Oral Pathol 64:146155 8. Loh FC (1992) Technical modification the sagittal split mandibular ramus osteotomy. Oral Surg Oral Med Oral Pathol 74:723726 9. Marquez IM, Stella JP (1998) Modification of sagittal split ramus osteotomy to avoid unfavorable fracture around impacted third molars. Int J Adult Orthod Orthognath Surg 13:183187 10. Teltzrow T, Kramer FJ, Schulze A, Baethge C, Brachvogel P (2005) Perioperative complications following sagittal split osteotomy of the mandible. J Craniomaxillofac Surg 33:307313 11. Guernsey LH, DeChamplain RW (1971) Sequelae and complications of the intra oral sagittal osteotomy in the mandibular rami. Oral Surg Oral Med Oral Pathol 32:176192 12. Behrman SJ (1972) Complications of sagittal osteotomy of the mandibular ramus. J Oral Surg 30:554561 13. Veras RB, Kriwalsky MS, Hoffmann S, Maurer P, Schubert J (2008) Functional and radiographic long-term results after bad split in orthognathic surgery. Int J Oral Maxillofac Surg 37:606611 14. Turvey TA (1985) Intraoperative complications of sagittal osteotomy of the mandibular ramus: incidence and management. J Oral Maxillofac Surg 43:504509 15. ORyan F (1990) Complications of orthognathic surgery. Oral Maxillofac Surg Clin North Am 2:593601 16. Panula K, Finne K, Oikarinen K (2001) Incidence of complications and problems related to orthognathic surgery: a review of 655 patients. J Oral Maxillofac Surg 59:11281137

Oral Maxillofac Surg (2012) 16:1927 17. Martis CS (1984) Complications of mandibular sagittal split osteotomy. J Oral Maxillofac Surg 42:101107 18. Akhtar S, Tuinzing DB (1999) Unfavorable splits in sagittal split osteotomy. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 87:267268 19. Simpson W (1981) Problems encountered in the sagittal split operation. Int J Oral Surg 10:8186 20. Epker BN, Fish LC (1986) Dentofacial Deformities: integrated Orthodontic and Surgical Correction. Vol. 1 St. Louis, CV Mosby, pp 232234 21. Mehra P, Castro V, Freitas RZ, Wolford LM (2001) Complications of the mandibular sagittal split ramus osteotomy associated with the presence or absence of third molars. J Oral Maxillofac Surg 59:854859 22. van Merkesteyn JP, Groot RH, van Leeuwaarden R, Kroon FH (1987) Intraoperative complications in sagittal and vertical ramus osteotomies. Int J Oral Maxillofac Surg 16:665670 23. el Deeb M, Wolford L, Bevis R (1989) Complications of orthognathic surgery. Clin Plast Surg 16:825840 24. Jnsson E, Svatrz K, Welander V (1979) Sagittal split technique. Part I. Immediate postoperative conditions: a radiographic follow-up study. Int J Oral Surg 8:7581 25. MacIntosh RB (1981) Experience with the sagittal osteotomy of the mandibular ramus: a 13-year review. J Maxillofac Surg 9:151165 26. Tucker MR (1995) Sagittal ramus osteotomy with and without third molars. J Oral Maxillofac Surg 53(suppl 4):80 27. Van de Perre JP, Stoelinga PJ, Blijdorp PA, Brouns JJ, Hoppenreijs TJ (1996) Perioperative morbidity in maxillofacial orthopaedic surgery: a retrospective study. J Craniomaxillofac Surg 24:263270 28. Precious DS, Lung KE, Pynn BR, Goodday RH (1998) Presence of impacted teeth as a determining factor of unfavorable splits in 1256 sagittal split osteotomies. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 85:362365 29. Acebal-Bianco F, Vuylsteke PL, Mommaerts MY, De Clercq CA (2000) Perioperative complications in corrective facial orthopedic surgery: a 5-year retrospective study. J Oral Maxillofac Surg 58:754760 30. Maurer P, Otto C, Eckert AW, Schubert J (2001) Komplikationen bei der chirurgischen Behandlung von Dysgnathien-ein 50 jhriger Behandlungsbericht. Mund Kiefer Gesichtschir 5:357361 31. Reyneke JP, Tsakiris P, Becker P (2002) Age as a factor in the complication rate after removal of unerupted/impacted third molars at the time of mandibular sagittal split osteotomy. J Oral Maxillofac Surg 60:654659 32. Kim SG, Park SS (2007) Incidence of complications and problems related to orthognathic surgery. J Oral Maxillofac Surg 65:24382444

27 33. Kriwalsky MS, Maurer P, Veras RB, Eckert AW, Schubert J (2008) Risk factors for a bad split during sagittal split osteotomy. Br J Oral Maxillofac Surg 46:177179 34. Falter B, Schepers S, Vrielinck L, Lambrichts I, Thijs H, Politis C (2010) Occurrence of bad splits during sagittal split osteotomy. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 110:430435 35. Schubert W, Kobienia BJ, Pollock RA (1997) Cross-sectional area of the mandible. J Oral Maxillofac Surg 55:689692, discussion 693 36. Schwartz H (2002) The timing of third molar removal in patients undergoing a bilateral sagittal split osteotomy. J Oral Maxillofac Surg 60:132133 37. Jones TA, Garg T, Monaghan A (2004) Removal of a deeply impacted mandibular third molar through a sagittal split ramus osteotomy approach. Br J Oral Maxillofac Surg 42:365368 38. Precious DS (2004) Removal of third molars with sagittal split osteotomies: the case for. J Oral Maxillofac Surg 62:11441146 39. Gil JN, Marin C, Claus JD, Lima SM Jr (2007) Modified osteotome for inferior border sagittal split osteotomy. J Oral Maxillofac Surg 65:18401842 40. Nagakawa K, Ueki K, Matsumoto N, Takatsuka S, Yamamoto E, Ooe H (1997) The assessment of trigeminal nerve paresthesia after bilateral sagittal split osteotomy: modified somatosensory evoked potentials recording methods. J Cranio Maxillofac Surg 25:97101 41. Joss CU, Vassalli IM (2009) Stability after bilateral sagittal split osteotomy advancement surgery with rigid internal fixation: a systematic review. J Oral Maxillofac Surg 67:301313 42. Tucker MR (2004) Prevention and management of bad splits in sagittal split osteotomies. J Oral Maxillofac Surg 62(suppl 1):14 43. Mommaerts MY (1992) Two similar bad splits and how they were treated. Report of two cases. Int J Oral Maxillofac Surg 21:331332 44. Patterson AL, Bagby SK (1999) Posterior vertical body osteotomy (PVBO): a predictable rescue procedure for proximal segment fracture during sagittal split ramus osteotomy of the mandible. J Oral Maxillofac Surg 57:475477 45. ORyan F, Poor DB (2004) Completing sagittal split osteotomy of the mandible after fracture of the buccal plate. J Oral Maxillofac Surg 62:11751176 46. Duguet V, Precious DS, Clinton R (1987) Clivage sagittal des branches montantes mandibulaires: prevention des tranmatismes du pedicule dentaire inferieur. Rev Stomatol Chir Maxillofac 88:7176 47. Simpson W (1972) The short lingual cut in the sagittal osteotomy. J Oral Surg 30:811812